Disease due to M. tuberculosis infections remains a global public health tragedy and a new, more effective vaccine is clearly needed to combat this devastating epidemic. In this project, we have been evaluating the safety and effectiveness of novel TB vaccines in a mouse model of pulmonary TB. Recently, we have examined the protective activity of TB DNA vaccines and live attenuated M. tuberculosis vaccine strains. We have demonstrated that a cocktail of TB DNA vaccines is as effective as the current vaccine, BCG, in the mouse aerosol infection model. We have also shown that specific TB DNA vaccines can boost BCG-induced immune responses using a prime-boost regimen. Surprisingly, immunization with the DNA vaccine cocktail is protective against tuberculous infection in CD4KO mice; this protective immunity is mediated largely by IFN-gamma producing CD8 cells. In the attenuated vaccine studies using immunocompetent mice, we have demonstrated that specific attenuated mutants and BCG induce similar levels of protective immunity. Importantly, one of these attenuated M. tuberculosis vaccine strains provides better protection and is safer than BCG when administered to immunocompromised animals. In future studies, we plan to continue evaluating novel TB vaccines in animal models and attempting to define the vaccine-mediated immune mechanisms responsible for the protective responses induced by these vaccines in mice. Persistence and immunogenicity of attenuated mutants of M. tuberculosis. The current tuberculosis vaccine, BCG, was generated by attenuating a mycobacterial strain (Mycobacterium bovis) that is related but not identical to M. tuberculosis. It is believed that an attenuated mutant of M. tb with its full complement of tuberculosis antigens would provide more protection than the modestly active BCG vaccine. To evaluate the potential effectiveness of attenuated M. Tb strains as Tb vaccines, virulent M. tb H37Rv has been attenuated by a targeted mutagenesis. The resulting gene deletion strains have been injected into C57BL/6 mice to assess their survival in the lung and spleen as a measure of residual virulence and potential immunogenicity. To assess the protection afforded by the attenuated strains, the mice have been challenged 6-12 weeks later with a small aerogenic inoculum of M. tb Erdman and the growth in the lung was compared with that of a group of BCG controls. In collaboration with Dr. William Jacobs of the Albert Einstein University, we have extensively evaluated 8 deletion mutants of M. Tb. Four strains have deletions in genes - Leu, Lys, NAD, and Pan - involved with intracellular metabolism. The RD1 mutant contains a large deletion in the RD1 region of the TB genome; this genetic locus contains important TB virulence determinants. Additionally, we have recently studied two multiple gene deletion strains of M. tuberculosis - the RD1pan and lyspan mutants. A summary of the results generated in these studies is provided below. 1. The auxotrophic H37RvLeu- mutant induced high levels of anti-tuberculous resistance but virulence tests carried out on this auxotroph indicated an unacceptable level of reversion to the wild type and further studies with this auxotroph were discontinued. 2. Tests with lysine auxotrophs of BCG Pasteur and M. tb H37Rv have been carried out in C57BL/6 mice. Both auxotrophs failed to multiply in vitro in the absence of an extrinsic source of lysine and were unable to grow or survive in intravenously vaccinated mice. They were unable to induce detectable anti-tuberculous resistance against an aerogenic Erdman challenge administered 3 months later. However, mice which received 2 or 3 doses of the lysine auxotroph expressed enhanced resistance to challenge, approaching that observed in the BCG vaccinated controls. 3. The NAD mutant persists for many months in C57BL/6 mice. Importantly, the challenge studies have indicated that a single dose of this strain is more protective than BCG. However, our virulence studies have indicated that this strain is not fully attenuated and hence would be unacceptable for clinical use. Because the NAD strain is persistent and immunogenic, an additional gene is being deleted from this strain to reduce its virulence. We anticipate that a NAD double KO strain will be sufficiently immunogenic and less toxic and a potential vaccine candidate. 4. The PAN mutant has a deletion in a gene responsible for the production of pantothenic acid. This mutant is only modestly persistent in our mouse model. In fact, when delivered subcutanously only a few organisms are detected in the spleens and lungs at 1 and 2 months post infection. Because of this modest persistence, the toxicity of this strain is limited. Surprisingly, the protective response induced by two subcutaneously injections of this mutant is equivalent to the BCG response. Thus, the PAN mutant is a viable vaccine candidate because of its low toxicity and substantial immongenicity. 5. The RD1 mutant carries a large deletion in a region of the TB genome encoding potential virulence determinants. This mutant persists in mice for months and protects as well as BCG against an aerogenic challenge. However, studies in immune deficient mice suggest that the RD1 mutant is not fully attenuated. 6. The RD1pan multiple gene deletion strain may be a viable vaccine candidate. This strain persists in mice at a low level for several months and is as protective as BCG in immunocompetent mice. Furthermore, studies in scid mice show that the RD1pan mutant is more attenuated than BCG in this mouse model. Importantly, the RD1pan mutant is significantly more protective than BCG in CD4KO mice. 7. Our studies with the Lyspan multiple deletion strain indicates that it is immunogenic and highly attenuated when it is administered to immunocompetent and immunocompromised mice. Characterization of Mycobacterial DNA vaccines. Tuberculosis is among the most important global public health concerns with more than 2 million people dying from this disease each year. The current tuberculosis vaccine, BCG, has questionable efficacy and is counterindicated in some of the target populations. Recent studies have indicated that DNA vaccines may be promising as new mycobacterial vaccines candidates. We have created 35 DNA vaccines expressing different mycobacterial antigens. These tuberculosis DNA vaccines have been generated to express TPA-fused antigens, ubiquitin-conjugated proteins and native antigens. All of these vaccines have been tested by intramuscular inoculation,either singly or in combination, for their capacity to induce protective immune responses in mice. Based on the results of these studies, we have concluded the following: 1. At least 12 tuberculosis DNA vaccines that we have tested induce protective immune responses in murine primary aerogenic infection models. However, the level of protection for single vaccines is less than the level of protection evoked by BCG. 2. Tuberculosis DNA vaccines encoding mycobacterial antigens fused to a TPA signal signal are more immunogenic than DNA vaccines encoding the native protein. TB DNA vaccines encoding mycobacterial antigens fused to ubiquitin induce primarily high levels of cell-mediated immunity and low humoral responses. 3. Combinations of TB DNA vaccines evoke a larger protective response than single vaccines. The level of protective immunity elicited by a 8 component TPA TB DNA vaccines combination and a 8 component ubiquitin DNA vaccine combination was equivalent to the BCG response in long-term studies. In three studies, mice immunized with either the TPA TB DNA vaccine combination, the ubiquitin vaccine cocktail, or BCG survived as much as 8-fold longer than naive mice after an aerogenic challenge with virulent M. tuberculosis. 4. The effectiveness of TB DNA vaccination has also been evaluated in a mouse model of latent tuberculosis. Immunization with the DNA vaccine combinations does not prevent reactivational disease in this model but does have a modest effect on reducing the growth of a secondary challenge (exogenous reinfection). 5. TB DNA vaccine cocktails can protect highly susceptible CD4-KO mice against a tuberculous challenge. The protective immunity induced by the DNA vaccines in this model is largely mediated by IFN-gamma producing CD8 cells. In contrast, our DNA vaccine combinations do not protect CD8-KO mice against an aerogenic M. tuberculosis infection. 6. TB DNA vaccines which contain a Sindbis virus replicon are highly immunogenic. Sindbis-based TB DNA vaccines are about 50-fold (per microgram) more active than conventional DNA vaccines. 7. Specific DNA vaccines can boost BCG responses using prime-boost regimens. Antibiotic resistance in mycobacteria. 1. In collaboration with Dr. Robert Reynolds of the Southern Research Institute, we are trying to identify the intracellular targets for 2 very active novel antituberculosis agents. When the drug targets are identified, we will be involved in the design of even better antituberculosis compounds using molecular modeling and combitorial chemistry techniques. The general approach is to generate resistant clones, create gene libraries from chromosomal DNA from these clones, transform drug sensitive cells with these libraries, and then search for resistance genes by plating transformants on drug selection media. For the initial studies, we have chosen to evaluate drug resistance in M. smegmatis because this mycobacterial strain is much easier and safer to work with than M. tuberculosis. Thus far, we have isolated several M. smegmatis clones that are highly resistant to the two pyrido-pyrazine carbamic acid compounds. We have also isolated DNA from resistant organisms and have created plasmid libraries. We are in the process of transforming drug sensitive cells with the resistance libraries and selecting for drug resistant transformants. 2. In collaboration with Dr. Larry Bockstahler from FDA/CDRH, we have been attempting to develop rapid and sensitive molecular methods to detect drug-resistant M. tuberculosis organisms. These methods are based on the identification of specific mutations in genes known to be associated with drug resistance in M. tuberculosis. Our initial efforts involved using peptide nucleic acid (PNA) probes to detect gene mutations. PNA probes are useful for detecting mutant nucleic acids, because they have high thermal stability, strong binding capacity and high binding specificity. Using a PNA-PCR-ELISA format, we have shown that we can quickly and specifically identify mutations in the M. tuberculosis katG (isoniazid resistance) and rpoB genes (rifampin resistance)that are known to be associated with drug resistance. Recently, we have extended this approach for quickly identifying TB mutations to a microarray format. We have shown that certain mutations in the katG and rpoB genes can be identified using the microarray technology. We are in the process of establishing the conditions to identify other mutations in TB genes associated with drug resistance. This project incorporates FY2002 projects 1Z01BJ006015-07, 1Z01BJ006018-05, and 1Z01BJ006021-02.